CN111184948A - Vascular targeted photodynamic therapy-based nevus flammeus treatment method and system - Google Patents
Vascular targeted photodynamic therapy-based nevus flammeus treatment method and system Download PDFInfo
- Publication number
- CN111184948A CN111184948A CN202010021564.9A CN202010021564A CN111184948A CN 111184948 A CN111184948 A CN 111184948A CN 202010021564 A CN202010021564 A CN 202010021564A CN 111184948 A CN111184948 A CN 111184948A
- Authority
- CN
- China
- Prior art keywords
- data
- model
- patient
- treatment
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N5/00—Radiation therapy
- A61N5/06—Radiation therapy using light
- A61N5/0613—Apparatus adapted for a specific treatment
- A61N5/062—Photodynamic therapy, i.e. excitation of an agent
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N5/00—Radiation therapy
- A61N5/06—Radiation therapy using light
- A61N5/067—Radiation therapy using light using laser light
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N5/00—Radiation therapy
- A61N5/06—Radiation therapy using light
- A61N2005/0626—Monitoring, verifying, controlling systems and methods
Landscapes
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Biophysics (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
By the aid of the method and the system for treating the nevus flammeus based on the vascular targeted photodynamic therapy, a predicted treatment scheme is more accurate, a training model is more complete, the weight of attribute data can be updated more scientifically and effectively, and a global optimal solution of a target function can be found well under certain conditions. The method comprises the following steps: (1) constructing a special disease database and desensitizing the data based on the case data of the past nevus flammeus patients; (2) constructing a personalized treatment scheme recommendation algorithm expert system based on a machine learning model by adopting a mode of matching a supervised model with an unsupervised model; (3) reading basic information of a patient and focus information of the patient which are required to be diagnosed at present; (4) performing data matching, and selecting a plurality of past cases which are most matched with the data as recommendations of similar cases; (5) carrying out regression analysis on the input data to obtain the prediction treatment scheme recommendation of the current patient; (6) and outputting the result.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to a nevus flammeus treatment method based on a blood vessel targeted photodynamic therapy and a nevus flammeus treatment system based on the blood vessel targeted photodynamic therapy.
Background
Port Wine Stain (PWS) is a common congenital capillary disease, which is frequently occurring in the face and neck, and the incidence rate in newborns is 0.3% -0.5%. At present, the total number of moles of nevus flammeus patients worldwide is estimated to be more than 2000 thousands of times, wherein the number of the existing patients in China is more than 600 thousands of times, the number of the existing patients is still continuously increased every year, and the clinical demand is large. The nevus flammeus generally cannot fade, and with the age, the color of the focus gradually deepens and thickens, even nodules and deformation appear, so that the appearance and the normal life of a patient are seriously affected, and serious psychological burden is brought to the patient.
The early selection of a proper treatment mode can reduce the treatment times, improve the curative effect, avoid and reduce the adverse effects of local skin thickening, tissue structure deformity, formation of suppurative granuloma and the like, and can also reduce the economic and psychological burden of patients. Vascular targeted Photodynamic Therapy (V-PDT) is the currently preferred method for treating the nevus flammeus, is most widely applied in China, and has the advantages of good selectivity, good applicability, repeatable treatment and the like.
However, the characteristics of the lesion of the nevus flammeus are complex, so that the skin damage of the nevus flammeus in clinical practice is different from person to person, and even if the same patient is in need, the lesion degree of the lesion at different positions is different. In the past 20 years, although V-PDT has achieved good clinical application effect, the difference of disease species, location, degree, individuals and other factors is often marked due to the individual difference of patients, and the regulation and control of the treatment dosage is lack of objective basis. This is because the therapeutic effect of V-PDT depends on the color of the lesion of nevus flammeus, the depth of the microvasculature, the diameter of the vessel, etc. How to conduct individualized quantitative evaluation according to individual differences of patients so as to guide the formulation of individualized V-PDT treatment dose and realize reasonable regulation, curative effect quantitative evaluation and the like becomes a challenging problem to be solved urgently. Therefore, establishing a standardized system has great significance for diagnosing, evaluating and accurately treating the port wine stains. Therefore, the invention provides a vascular targeting photodynamic therapy-based individualized recommendation system for a nevus flammeus treatment scheme, which is used for accurately controlling the treatment dosage and assisting doctors in diagnosis and treatment.
A typical personalized treatment recommendation system usually assigns a different weight to different data attributes (e.g. age, height, etc. of a patient) to measure the importance of the data attributes to a treatment plan. The method for determining the weight value simply is based on experience information of experts in related fields, and because the experience of each expert is different, each expert also has individual difference, and because the experience information is difficult to quantize into a specific numerical value, the effect of the method is often poor. Another method for determining the weight is based on direct optimization of some specified objective function, which is often difficult to optimize or only find a local optimal solution of the optimization objective due to the complex relationship between the objective function and the data attribute.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a method for treating nevus flammeus based on a vascular targeted photodynamic therapy, the predicted treatment scheme is more accurate, the training model is more complete, the weight of attribute data can be more scientifically and effectively updated, and the global optimal solution of the objective function can be well found under certain conditions.
The technical scheme of the invention is as follows: the method for treating port wine stains based on the vascular targeted photodynamic therapy comprises the following steps:
(1) establishing a database, constructing a special disease database based on the case data of the past nevus flammeus patients, and desensitizing the data;
(2) establishing an expert system model, constructing a personalized treatment scheme recommendation algorithm expert system based on a machine learning model by adopting a mode of matching a supervised model and an unsupervised model, acquiring case data in a special disease database to train the expert system, recommending a treatment scheme to a patient by the trained expert system, and continuously training the model on line so as to be more perfect;
(3) establishing input data, and reading basic information of a patient and focus information of the patient which need diagnosis and treatment at present;
(4) performing a first part of prediction, performing data matching on the patient data in the step (3) and a special disease database of the nevus flammeus by an unsupervised machine learning model, selecting a plurality of past cases which are most matched with the patient data and the special disease database of the nevus flammeus as recommendations of similar cases, and taking the recommendations as a part of a final prediction result;
(5) performing a second part of prediction, taking the data of the current patient in the step (3) as input, and performing regression analysis on the input data through the model trained in the step (2) to obtain the prediction treatment scheme recommendation of the current patient;
(6) and (4) outputting a result, namely performing weighting processing on the predicted output of the supervised model in the step (5) and the treatment scheme of the unsupervised model in the step (4), and taking the result as the final output of the personalized treatment scheme recommendation system.
The invention weights the historical treatment scheme similar to the current patient condition and the treatment scheme predicted by the machine learning model trained based on the special disease database, and the weighting mode of the invention has larger weight to the treatment scheme which is extremely similar to the current patient in the historical data, and certain interpretability is reserved under the conditions that the predicted treatment scheme is more accurate and the training model is more perfect; the machine learning model in the expert system can update the weight of the attribute data more scientifically and effectively by an optimization method based on gradient descent, and can well find the global optimal solution of the target function under certain conditions.
Also provided is a nevus flammeus treatment system based on vascular targeted photodynamic therapy, comprising:
the database establishing module is used for establishing a database, establishing a special disease database based on the case data of the past nevus flammeus patients and desensitizing the data;
the expert system model building module builds an expert system model, adopts a mode of matching a supervised model and an unsupervised model to build a personalized treatment scheme recommendation algorithm expert system based on a machine learning model, collects case data in a special disease database to train the expert system, and the trained expert system recommends a treatment scheme for a patient while the model continues on-line training to be more perfect;
the input data establishing module is used for establishing input data and reading basic information of a patient and focus information of the patient which need diagnosis and treatment at present;
the first part prediction module is used for performing the first part prediction, performing data matching on the disease data and a special disease database of the nevus flammeus by an unsupervised machine learning model, selecting a plurality of past cases which are most matched with the disease data as recommendations of similar cases, and using the recommendations as a part of a final prediction result;
the second part prediction module is used for performing second part prediction, taking the data of the current patient as input, and performing regression analysis on the input data through the trained model to obtain the prediction treatment scheme recommendation of the current patient;
and the output module outputs a result, performs weighting processing on the prediction output of the supervised model and the treatment scheme of the unsupervised model, and takes the result as the final output of the personalized treatment scheme recommendation system.
Drawings
Fig. 1 is a flow chart of a nevus flammeus treatment method based on vascular targeted photodynamic therapy according to the present invention.
Detailed Description
As shown in fig. 1, the method for treating port wine stains based on vascular targeted photodynamic therapy comprises the following steps:
(1) establishing a database, constructing a special disease database based on the case data of the past nevus flammeus patients, and desensitizing the data;
(2) establishing an expert system model, constructing a personalized treatment scheme recommendation algorithm expert system based on a machine learning model by adopting a mode of matching a supervised model and an unsupervised model, acquiring case data in a special disease database to train the expert system, recommending a treatment scheme to a patient by the trained expert system, and continuously training the model on line so as to be more perfect;
(3) establishing input data, and reading basic information of a patient and focus information of the patient which need diagnosis and treatment at present;
(4) performing a first part of prediction, performing data matching on the patient data in the step (3) and a special disease database of the nevus flammeus by an unsupervised machine learning model, selecting a plurality of past cases which are most matched with the patient data and the special disease database of the nevus flammeus as recommendations of similar cases, and taking the recommendations as a part of a final prediction result;
(5) performing a second part of prediction, taking the data of the current patient in the step (3) as input, and performing regression analysis on the input data through the model trained in the step (2) to obtain the prediction treatment scheme recommendation of the current patient;
(6) and (4) outputting a result, namely performing weighting processing on the predicted output of the supervised model in the step (5) and the treatment scheme of the unsupervised model in the step (4), and taking the result as the final output of the personalized treatment scheme recommendation system.
The invention weights the historical treatment scheme similar to the current patient condition and the treatment scheme predicted by the machine learning model trained based on the special disease database, and the weighting mode of the invention has larger weight to the treatment scheme which is extremely similar to the current patient in the historical data, and certain interpretability is reserved under the conditions that the predicted treatment scheme is more accurate and the training model is more complete; the machine learning model in the expert system can update the weight of the attribute data more scientifically and effectively by an optimization method based on gradient descent, and can well find the global optimal solution of the target function under certain conditions.
Preferably, the step (2) comprises the following substeps:
(2.1) extracting data in the special disease database and performing data cleaning: removing samples with abnormal statistics; filling features with few missing values in a mode of mode and average value; observing whether the data are balanced or not, and properly weighting or ignoring a certain characteristic value when a sample number of the characteristic value is less; normalizing or standardizing the continuous data; performing box separation processing on the discrete data; performing exponential or logarithmic transformation on the special data;
(2.2) performing characteristic engineering on the data: generating new characteristics according to case information logic; if the characteristic dimension is higher, performing characteristic dimension reduction by adopting a PCA matrix decomposition algorithm; selecting the features by adopting a Filter, Wrapper or Embedded mode, calculating the relevance between different feature values, mutual information and implicit relation between related distance indexes and measurement features, or calculating the weight of the features to a predicted value by adopting a regression model so as to calibrate the relative importance of the features, and finally determining the features needing to be trained.
Preferably, in the step (3),
the basic information and focus information data of the current patient are used as historical data input into an X and nevus flammeus databaseSimilarity matching is carried out through a similarity model, a Pearson correlation coefficient rho is used as similarity measurement among case samples, a threshold tau is designated, the value which is most matched with the characteristic value of the current patient is selected, the Pearson correlation coefficient among the case samples is larger than or equal to the threshold,n historical cases; arranging historical cases according to the descending order of the Pearson correlation coefficients rho, and enabling the respective correlation coefficients to be rho1,ρ2...ρn(ii) a Then each selected similar historical caseActual therapeutic doseCorresponding weight valueComprises the following steps:
where λ is 0.9 as attenuation factor and weightThe formula (c) shows that the influence on model prediction is smaller for the case with smaller correlation coefficient, and the final weighting resultAs part of the final treatment regimen recommendation.
Preferably, in the step (4), the historical data X in the disease database corresponding to the features in the step (2) is used as input data, the input data is input into a supervised machine learning model for training, and more than 5 base learners are simultaneously trained based on an integrated learning method in machine learning, wherein the base learners include random forest RF, decision tree DT, gradient boosting tree GBDT, extreme random forest ETR and neural network ANNs (shallow layer); after the base learner is trained, weighted averaging, voting or training of the secondary learner again by adopting a Stacking method is carried out. After the integrated learning model finishes training the historical data, the integrated learning model finally predicts the current patient data YESAs another part of the final treatment regimen recommendation.
Preferably, in the step (5), the treatment plans output in the steps (3) and (4) are weighted, and the predicted treatment plan weight in the step (4) isThe final recommended treatment is Y ═ YH+ωESYES。
Preferably, the method further comprises the steps of:
(7) desensitizing the complete case information in the step (6) and then returning the desensitized case information to a special disease database of port wine stains;
(8) and performing incremental learning on the expert system model to finish the online training of the expert system.
It will be understood by those skilled in the art that all or part of the steps in the method of the above embodiments may be implemented by hardware instructions related to a program, the program may be stored in a computer-readable storage medium, and when executed, the program includes the steps of the method of the above embodiments, and the storage medium may be: ROM/RAM, magnetic disks, optical disks, memory cards, and the like. Therefore, corresponding to the method of the present invention, the present invention also includes a nevus flammeus treatment system based on vascular targeted photodynamic therapy, which is generally expressed in the form of functional modules corresponding to the steps of the method.
The system comprises:
the database establishing module is used for establishing a database, establishing a special disease database based on the case data of the past nevus flammeus patients and desensitizing the data;
the expert system model building module builds an expert system model, adopts a mode of matching a supervised model and an unsupervised model to build a personalized treatment scheme recommendation algorithm expert system based on a machine learning model, collects case data in a special disease database to train the expert system, and the trained expert system recommends a treatment scheme for a patient while the model continues on-line training to be more perfect;
the input data establishing module is used for establishing input data and reading basic information of a patient and focus information of the patient which need diagnosis and treatment at present;
the first part prediction module is used for performing the first part prediction, performing data matching on the disease data and a special disease database of the nevus flammeus by an unsupervised machine learning model, selecting a plurality of past cases which are most matched with the disease data as recommendations of similar cases, and using the recommendations as a part of a final prediction result;
the second part prediction module is used for performing second part prediction, taking the data of the current patient as input, and performing regression analysis on the input data through the trained model to obtain the prediction treatment scheme recommendation of the current patient;
and the output module outputs a result, performs weighting processing on the prediction output of the supervised model and the treatment scheme of the unsupervised model, and takes the result as the final output of the personalized treatment scheme recommendation system.
Compared with the existing personalized treatment recommendation system, the method has the advantages that:
1. weighting the historical treatment scheme similar to the current patient condition and the treatment scheme predicted by the machine learning model trained based on the special disease database to ensure that the recommended scheme is more accurate, and meanwhile, the weighting mode has higher weight on the treatment scheme which is extremely similar to the current patient in the historical data, so that the model has certain interpretability;
2. after the treatment scheme of the current patient is determined, the current patient can be stored in a patient-specific database and an expert system is trained on line, and the model is more perfect along with the increase of the database;
3. the machine learning model in the expert system can update the weight of the attribute data more scientifically and effectively by an optimization method based on gradient descent, and can well find the global optimal solution of the target function under certain conditions.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and all simple modifications, equivalent variations and modifications made to the above embodiment according to the technical spirit of the present invention still belong to the protection scope of the technical solution of the present invention.
Claims (7)
1. A method for treating port wine stains based on vascular targeted photodynamic therapy is characterized in that: which comprises the following steps:
(1) establishing a database, constructing a special disease database based on the case data of the past nevus flammeus patients, and desensitizing the data;
(2) establishing an expert system model, constructing a personalized treatment scheme recommendation algorithm expert system based on a machine learning model by adopting a mode of matching a supervised model and an unsupervised model, acquiring case data in a special disease database to train the expert system, recommending a treatment scheme to a patient by the trained expert system, and continuously training the model on line so as to be more perfect;
(3) establishing input data, and reading basic information of a patient and focus information of the patient which need diagnosis and treatment at present;
(4) performing a first part of prediction, performing data matching on the patient data in the step (3) and a special disease database of the nevus flammeus by an unsupervised machine learning model, selecting a plurality of past cases which are most matched with the patient data and the special disease database of the nevus flammeus as recommendations of similar cases, and taking the recommendations as a part of a final prediction result;
(5) performing a second part of prediction, taking the data of the current patient in the step (3) as input, and performing regression analysis on the input data through the model trained in the step (2) to obtain the prediction treatment scheme recommendation of the current patient;
(6) and (4) outputting a result, namely performing weighting processing on the predicted output of the supervised model in the step (5) and the treatment scheme of the unsupervised model in the step (4), and taking the result as the final output of the personalized treatment scheme recommendation system.
2. The nevus flammeus treatment method based on vascular targeted photodynamic therapy according to claim 1, wherein: the step (2) comprises the following sub-steps:
(2.1) extracting data in the special disease database and performing data cleaning: removing samples with abnormal statistics; filling features with few missing values in a mode of mode and average value; observing whether the data are balanced or not, and properly weighting or ignoring a certain characteristic value when a sample number of the characteristic value is less; normalizing or standardizing the continuous data; performing box separation processing on the discrete data; performing exponential or logarithmic transformation on the special data;
(2.2) performing characteristic engineering on the data: generating new characteristics according to case information logic; if the characteristic dimension is higher, performing characteristic dimension reduction by adopting a PCA matrix decomposition algorithm; selecting the features by adopting a Filter, Wrapper or Embedded mode, calculating the relevance between different feature values, mutual information and implicit relation between related distance indexes and measurement features, or calculating the weight of the features to a predicted value by adopting a regression model so as to calibrate the relative importance of the features, and finally determining the features needing to be trained.
3. The nevus flammeus treatment method based on vascular targeted photodynamic therapy according to claim 2, wherein: in the step (3), the step (c),
the basic information and focus information data of the current patient are used as historical data input into an X and nevus flammeus databaseSimilarity matching is carried out through a similarity model, a Pearson correlation coefficient rho is used as similarity measurement among case samples, a threshold tau is designated, the value which is most matched with the characteristic value of the current patient is selected, the Pearson correlation coefficient among the case samples is larger than or equal to the threshold,n historical cases; arranging historical cases according to the descending order of the Pearson correlation coefficients rho, and enabling the respective correlation coefficients to be rho1,ρ2...ρn(ii) a Then each selected similar historical caseActual therapeutic doseCorresponding weight valueComprises the following steps:
4. The nevus flammeus treatment method based on vascular targeted photodynamic therapy according to claim 3, wherein: in the step (4), historical data X in the special disease database corresponding to the characteristics in the step (2) is used as input data and input into a supervised machine learning model for training, and more than 5 base learners are simultaneously trained based on an integrated learning method in machine learning, wherein the base learners comprise random forest RF, decision tree DT, gradient boosting tree GBDT, extreme random forest ETR and neural network ANNs (shallow layer); after the base learner is trained, weighted averaging, voting or training of the secondary learner again by adopting a Stacking method is carried out. After the integrated learning model finishes training the historical data, the integrated learning model finally predicts the current patient data YESAs another part of the final treatment regimen recommendation.
5. The nevus flammeus treatment method based on vascular targeted photodynamic therapy according to claim 4, wherein: in the step (5), the treatment schemes output in the step (3) and the step (4) are weighted, and the predicted treatment scheme weight in the step (4) isThe final recommended treatment is Y ═ YH+ωESYES。
6. The nevus flammeus treatment method based on vascular targeted photodynamic therapy according to claim 1, wherein: the method further comprises the following steps:
(7) desensitizing the complete case information in the step (6) and then returning the desensitized case information to a special disease database of port wine stains;
(8) and performing incremental learning on the expert system model to finish the online training of the expert system.
7. Vascular targeting photodynamic therapy based on bright red nevus processing system, its characterized in that: it includes:
the database establishing module is used for establishing a database, establishing a special disease database based on the case data of the past nevus flammeus patients and desensitizing the data;
the expert system model building module builds an expert system model, adopts a mode of matching a supervised model and an unsupervised model to build a personalized treatment scheme recommendation algorithm expert system based on a machine learning model, collects case data in a special disease database to train the expert system, and the trained expert system recommends a treatment scheme for a patient while the model continues on-line training to be more perfect;
the input data establishing module is used for establishing input data and reading basic information of a patient and focus information of the patient which need diagnosis and treatment at present;
the first part prediction module is used for performing the first part prediction, performing data matching on the disease data and a special disease database of the nevus flammeus by an unsupervised machine learning model, selecting a plurality of past cases which are most matched with the disease data as recommendations of similar cases, and using the recommendations as a part of a final prediction result;
the second part prediction module is used for performing second part prediction, taking the data of the current patient as input, and performing regression analysis on the input data through the trained model to obtain the prediction treatment scheme recommendation of the current patient;
and the output module outputs a result, performs weighting processing on the prediction output of the supervised model and the treatment scheme of the unsupervised model, and takes the result as the final output of the personalized treatment scheme recommendation system.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010021564.9A CN111184948B (en) | 2020-01-09 | 2020-01-09 | Vascular targeted photodynamic therapy-based nevus flammeus treatment method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010021564.9A CN111184948B (en) | 2020-01-09 | 2020-01-09 | Vascular targeted photodynamic therapy-based nevus flammeus treatment method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111184948A true CN111184948A (en) | 2020-05-22 |
CN111184948B CN111184948B (en) | 2020-12-25 |
Family
ID=70684832
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010021564.9A Active CN111184948B (en) | 2020-01-09 | 2020-01-09 | Vascular targeted photodynamic therapy-based nevus flammeus treatment method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111184948B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113221162A (en) * | 2021-04-28 | 2021-08-06 | 健康数据(北京)科技有限公司 | Private disease-specific big data privacy protection method and system based on block chain |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101670152A (en) * | 2009-09-10 | 2010-03-17 | 北京理工大学 | Photodynamic therapy system |
CN109036568A (en) * | 2018-09-03 | 2018-12-18 | 浪潮软件集团有限公司 | Method for establishing prediction model based on naive Bayes algorithm |
WO2019060882A1 (en) * | 2017-09-25 | 2019-03-28 | Letterie Gerard | System for supporting clinical decision-making in reproductive endocrinology and infertility |
CN106897545B (en) * | 2017-01-05 | 2019-04-30 | 浙江大学 | A kind of tumor prognosis forecasting system based on depth confidence network |
CN110197724A (en) * | 2019-03-12 | 2019-09-03 | 平安科技(深圳)有限公司 | Predict the method, apparatus and computer equipment in diabetes illness stage |
CN110297179A (en) * | 2018-05-11 | 2019-10-01 | 宫文峰 | Diesel-driven generator failure predication and monitoring system device based on integrated deep learning |
CN110534190A (en) * | 2018-05-24 | 2019-12-03 | 西门子医疗有限公司 | System and method for automatic Clinical Decision Support Systems |
CN110556178A (en) * | 2018-05-30 | 2019-12-10 | 西门子医疗有限公司 | decision support system for medical therapy planning |
-
2020
- 2020-01-09 CN CN202010021564.9A patent/CN111184948B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101670152A (en) * | 2009-09-10 | 2010-03-17 | 北京理工大学 | Photodynamic therapy system |
CN106897545B (en) * | 2017-01-05 | 2019-04-30 | 浙江大学 | A kind of tumor prognosis forecasting system based on depth confidence network |
WO2019060882A1 (en) * | 2017-09-25 | 2019-03-28 | Letterie Gerard | System for supporting clinical decision-making in reproductive endocrinology and infertility |
CN110297179A (en) * | 2018-05-11 | 2019-10-01 | 宫文峰 | Diesel-driven generator failure predication and monitoring system device based on integrated deep learning |
CN110534190A (en) * | 2018-05-24 | 2019-12-03 | 西门子医疗有限公司 | System and method for automatic Clinical Decision Support Systems |
CN110556178A (en) * | 2018-05-30 | 2019-12-10 | 西门子医疗有限公司 | decision support system for medical therapy planning |
CN109036568A (en) * | 2018-09-03 | 2018-12-18 | 浪潮软件集团有限公司 | Method for establishing prediction model based on naive Bayes algorithm |
CN110197724A (en) * | 2019-03-12 | 2019-09-03 | 平安科技(深圳)有限公司 | Predict the method, apparatus and computer equipment in diabetes illness stage |
Non-Patent Citations (5)
Title |
---|
周扬 等: "《光动力疗法治疗鲜红斑痣的研究进展》", 《中国激光医学杂志》 * |
王乐天 等: "《基于数据库的光动力治疗鲜红斑痣专家***》", 《中国激光医学杂志》 * |
穆锦荣 等: "《基于监督学习半自动检测鲜红斑痣病灶的方法》", 《中国激光医学杂志》 * |
蒋孝良 等: "《贝叶斯网络分类器算法与应用》", 31 October 2015 * |
顾瑛: "《微血管靶向光动力疗法治疗鲜红斑痣》", 《第十五次全国皮肤性病学术年会》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113221162A (en) * | 2021-04-28 | 2021-08-06 | 健康数据(北京)科技有限公司 | Private disease-specific big data privacy protection method and system based on block chain |
Also Published As
Publication number | Publication date |
---|---|
CN111184948B (en) | 2020-12-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111009321A (en) | Application method of machine learning classification model in juvenile autism auxiliary diagnosis | |
CN110974214A (en) | Automatic electrocardiogram classification method, system and equipment based on deep learning | |
Taylor et al. | A model to detect heart disease using machine learning algorithm | |
CN113077887B (en) | Automatic quantitative analysis system and interpretation method for white matter lesions of brain | |
van der Westhuizen et al. | Bayesian LSTMs in medicine | |
CN116864139A (en) | Disease risk assessment method, device, computer equipment and readable storage medium | |
CN111184948B (en) | Vascular targeted photodynamic therapy-based nevus flammeus treatment method and system | |
Foomani et al. | Synthesizing time-series wound prognosis factors from electronic medical records using generative adversarial networks | |
CN116864104A (en) | Chronic thromboembolic pulmonary artery high-pressure risk classification system based on artificial intelligence | |
US20220223231A1 (en) | Systems and Methods for Improved Prognostics in Medical Imaging | |
Wein et al. | Forecasting brain activity based on models of spatiotemporal brain dynamics: A comparison of graph neural network architectures | |
Sauty et al. | Riemannian metric learning for progression modeling of longitudinal datasets | |
Herath et al. | Autism spectrum disorder diagnosis support model using Inception V3 | |
Bordin et al. | Explainable AI points to white matter hyperintensities for Alzheimer's disease identification: A preliminary study | |
Siddiqui et al. | Trust metrics for medical deep learning using explainable-ai ensemble for time series classification | |
CN112259232B (en) | VTE risk automatic evaluation system based on deep learning | |
Chen et al. | Estimating brain functional networks based on adaptively-weighted fMRI signals for MCI identification | |
US20230154583A1 (en) | Means for categorizing and treating disordered situations | |
US20230162827A1 (en) | Method and system for predicting neurological treatment | |
CN112329921B (en) | Diuretic dose reasoning equipment based on deep characterization learning and reinforcement learning | |
Kleinbaum et al. | Recurrent event survival analysis | |
CN115337000A (en) | Machine learning method for evaluating brain aging caused by diseases based on brain structure images | |
Shetty et al. | Alzheimer’s Diseases Detection by using Convolution Neural Network | |
Soguero-Ruiz et al. | Finding associations among chronic conditions by bootstrap and multiple correspondence analysis | |
US20220051809A1 (en) | System of predicting risks with biomedical data, method of predicting risks with biomedical data, and non-transient state computer-readable storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |